create_dl_layer_loss_distanceT_create_dl_layer_loss_distanceCreateDlLayerLossDistanceCreateDlLayerLossDistancecreate_dl_layer_loss_distance (Operator)

Name

create_dl_layer_loss_distanceT_create_dl_layer_loss_distanceCreateDlLayerLossDistanceCreateDlLayerLossDistancecreate_dl_layer_loss_distance — Create a distance loss layer.

Signature

Description

The operator create_dl_layer_loss_distancecreate_dl_layer_loss_distanceCreateDlLayerLossDistanceCreateDlLayerLossDistancecreate_dl_layer_loss_distance creates a distance loss layer whose handle is returned in DLLayerLossDistanceDLLayerLossDistanceDLLayerLossDistanceDLLayerLossDistancedllayer_loss_distance.

This layer expects multiple layers as input:

• DLLayerInputDLLayerInputDLLayerInputDLLayerInputdllayer_input: Specifies the prediction (e.g., a softmax layer).

• DLLayerTargetDLLayerTargetDLLayerTargetDLLayerTargetdllayer_target: Specifies the target sequences (originating from the ground truth information).

• DLLayerWeightsDLLayerWeightsDLLayerWeightsDLLayerWeightsdllayer_weights: Specifies the weight sequences. This parameter is optional. If an empty tuple [] is passed for all values the weighting factor 1.0 is used.

The parameter LayerNameLayerNameLayerNamelayerNamelayer_name sets an individual layer name. Note that if creating a model using create_dl_modelcreate_dl_modelCreateDlModelCreateDlModelcreate_dl_model each layer of the created network must have a unique name.

The parameter LossWeightLossWeightLossWeightlossWeightloss_weight is an overall loss weight if there are multiple losses in the network.

The parameter DistanceTypeDistanceTypeDistanceTypedistanceTypedistance_type determines which distance measure is applied. Currently only 'l2'"l2""l2""l2""l2" is implemented. Depending on the generic parameter 'reduce'"reduce""reduce""reduce""reduce" this results in

• L2 loss distance as a tensor: in this case loss is the tensor of the same size as DLLayerInputDLLayerInputDLLayerInputDLLayerInputdllayer_input.

• L2 loss distance as a scalar: where is a number of elements in DLLayerInputDLLayerInputDLLayerInputDLLayerInputdllayer_input.

Thus DLLayerInputDLLayerInputDLLayerInputDLLayerInputdllayer_input, DLLayerTargetDLLayerTargetDLLayerTargetDLLayerTargetdllayer_target and DLLayerWeightsDLLayerWeightsDLLayerWeightsDLLayerWeightsdllayer_weights should have the same size. Setting the weights in DLLayerWeightsDLLayerWeightsDLLayerWeightsDLLayerWeightsdllayer_weights to 1 will result in a loss normalized over the number of elements.

The following generic parameters GenParamNameGenParamNameGenParamNamegenParamNamegen_param_name and the corresponding values GenParamValueGenParamValueGenParamValuegenParamValuegen_param_value are supported:

'is_inference_output'"is_inference_output""is_inference_output""is_inference_output""is_inference_output":

Determines whether apply_dl_modelapply_dl_modelApplyDlModelApplyDlModelapply_dl_model will include the output of this layer in the dictionary DLResultBatchDLResultBatchDLResultBatchDLResultBatchdlresult_batch even without specifying this layer in OutputsOutputsOutputsoutputsoutputs ('true'"true""true""true""true") or not ('false'"false""false""false""false").

Default: 'false'"false""false""false""false"

'reduce'"reduce""reduce""reduce""reduce":

Determines whether the output of the layer is reduced:

• 'true'"true""true""true""true": The output is reduced to a scalar.

• 'false'"false""false""false""false": The output of the layer is a tensor, where each element is a 'per-pixel' loss (squared differences).

Default: 'true'"true""true""true""true".

Certain parameters of layers created using this operator create_dl_layer_loss_distancecreate_dl_layer_loss_distanceCreateDlLayerLossDistanceCreateDlLayerLossDistancecreate_dl_layer_loss_distance can be set and retrieved using further operators. The following tables give an overview, which parameters can be set using set_dl_model_layer_paramset_dl_model_layer_paramSetDlModelLayerParamSetDlModelLayerParamset_dl_model_layer_param and which ones can be retrieved using get_dl_model_layer_paramget_dl_model_layer_paramGetDlModelLayerParamGetDlModelLayerParamget_dl_model_layer_param or get_dl_layer_paramget_dl_layer_paramGetDlLayerParamGetDlLayerParamget_dl_layer_param. Note, the operators set_dl_model_layer_paramset_dl_model_layer_paramSetDlModelLayerParamSetDlModelLayerParamset_dl_model_layer_param and get_dl_model_layer_paramget_dl_model_layer_paramGetDlModelLayerParamGetDlModelLayerParamget_dl_model_layer_param require a model created by create_dl_modelcreate_dl_modelCreateDlModelCreateDlModelcreate_dl_model.

Execution Information

• Multithreading type: reentrant (runs in parallel with non-exclusive operators).
• Multithreading scope: global (may be called from any thread).
• Processed without parallelization.

Parameters

Module

Deep Learning Training